This study will investigate the distribution and determinants of cancer occurrence in a defined population of homosexual men infected with human immuno-deficiency virus (HIV) or at high risk of HIV infection. Subjects will be the 4,955 homosexual men enrolled in the Multicenter AIDS Cohort Study (MACS) plus the supplemental cohort currently being enrolled. This population is drawn from four different regions of the U.S. and has been intensively followed for 3 1/2 years. In addition to extensive longitudinal questionnaire information regarding life style, drug use, sexual practices, past medical history, AIDS therapy, etc., a bank of biologic specimens (white cells, sera, semen, urine, saliva, etc.) collected at frequented intervals from these subjects has been maintained. Of this cohort nearly 2,000 were HIV-antibody positive at time of entry and approximately 270 of the initially seronegative subjects have sero-converted to HIV-positive. To date, 88 cancers have occurred in the total cohort. This study uses a multidisciplinary approach to determine the incidence of malignancy in HIV infected persons and to assess potential risk factors and risk markers for the development of these malignancies. The research team includes epidemiologists, biostatisticians, virologists, immunologists, and pathologists. Attempts will be made to identify etiologic mechanisms with particular emphasis on the roles of viruses and immune function. The design of the MACS allows for the performance of cohort study analyses for those variables on which data have been collected on all subjects (i.e., life style factors, T4/T8 ratios, etc.). Additionally, the banking of biologic specimens allows the use of nested case-control study analyses for assessing laboratory variables such as natural killer cell function and virologic studies. Tumor specimens from these subjects will be intensively studied at two of the MACS Center. Autopsies will be performed on subjects dying of AIDS to assess the prevalence of occult malignancies. A tumor bank will assembled and maintained for use in this and future research. Comparisons will be made between HIV-associated tumors and similar tumors occurring in the large transplantation population seen at the University of Pittsburgh. The data from this research will be merged with the other MACS data so that comprehensive analyses can be done.